Quantization made by Richard Erkhov.

multimaster-7b-v6 - GGUF

Model creator: https://huggingface.co/ibivibiv/
Original model: https://huggingface.co/ibivibiv/multimaster-7b-v6/

Name	Quant method	Size
multimaster-7b-v6.Q2_K.gguf	Q2_K	12.04GB
multimaster-7b-v6.IQ3_XS.gguf	IQ3_XS	13.48GB
multimaster-7b-v6.IQ3_S.gguf	IQ3_S	14.25GB
multimaster-7b-v6.Q3_K_S.gguf	Q3_K_S	14.23GB
multimaster-7b-v6.IQ3_M.gguf	IQ3_M	14.49GB
multimaster-7b-v6.Q3_K.gguf	Q3_K	15.79GB
multimaster-7b-v6.Q3_K_M.gguf	Q3_K_M	15.79GB
multimaster-7b-v6.Q3_K_L.gguf	Q3_K_L	17.1GB
multimaster-7b-v6.IQ4_XS.gguf	IQ4_XS	17.79GB
multimaster-7b-v6.Q4_0.gguf	Q4_0	18.6GB
multimaster-7b-v6.IQ4_NL.gguf	IQ4_NL	18.77GB
multimaster-7b-v6.Q4_K_S.gguf	Q4_K_S	18.76GB
multimaster-7b-v6.Q4_K.gguf	Q4_K	19.96GB
multimaster-7b-v6.Q4_K_M.gguf	Q4_K_M	19.96GB
multimaster-7b-v6.Q4_1.gguf	Q4_1	20.65GB
multimaster-7b-v6.Q5_0.gguf	Q5_0	22.7GB
multimaster-7b-v6.Q5_K_S.gguf	Q5_K_S	22.7GB
multimaster-7b-v6.Q5_K.gguf	Q5_K	23.41GB
multimaster-7b-v6.Q5_K_M.gguf	Q5_K_M	23.41GB
multimaster-7b-v6.Q5_1.gguf	Q5_1	24.76GB
multimaster-7b-v6.Q6_K.gguf	Q6_K	27.07GB
multimaster-7b-v6.Q8_0.gguf	Q8_0	35.06GB

Original model description:

language: - en license: apache-2.0 library_name: transformers model-index: - name: multimaster-7b-v6 results: - task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics: - type: acc_norm value: 72.78 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ibivibiv/multimaster-7b-v6 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics: - type: acc_norm value: 88.77 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ibivibiv/multimaster-7b-v6 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics: - type: acc value: 64.74 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ibivibiv/multimaster-7b-v6 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics: - type: mc2 value: 70.89 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ibivibiv/multimaster-7b-v6 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics: - type: acc value: 86.42 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ibivibiv/multimaster-7b-v6 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics: - type: acc value: 70.36 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ibivibiv/multimaster-7b-v6 name: Open LLM Leaderboard

Multi Master 7Bx5 v6

A quick multi-disciplinary moe model. This is part of a series of models built to test the gate tuning for mixtral style moe models.

Prompting

Prompt Template for alpaca style

### Instruction:

<prompt> (without the <>)

### Response:

Sample Code

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

torch.set_default_device("cuda")

model = AutoModelForCausalLM.from_pretrained("ibivibiv/multimaster-7b-v6", torch_dtype="auto", device_config='auto')
tokenizer = AutoTokenizer.from_pretrained("ibivibiv/multimaster-7b-v6")

inputs = tokenizer("### Instruction: Who would when in an arm wrestling match between Abraham Lincoln and Chuck Norris?\nA. Abraham Lincoln \nB. Chuck Norris\n### Response:\n", return_tensors="pt", return_attention_mask=False)

outputs = model.generate(**inputs, max_length=200)
text = tokenizer.batch_decode(outputs)[0]
print(text)

Model Details

Trained by: ibivibiv
Library: HuggingFace Transformers
Model type: multimaster-7b is a lora tuned version of openchat/openchat-3.5-0106 with the adapter merged back into the main model
Language(s): English
Purpose: This model is a focus on multi-disciplinary model tuning

Benchmark Scores

coming soon

Citations

@misc{open-llm-leaderboard,
  author = {Edward Beeching and Clémentine Fourrier and Nathan Habib and Sheon Han and Nathan Lambert and Nazneen Rajani and Omar Sanseviero and Lewis Tunstall and Thomas Wolf},
  title = {Open LLM Leaderboard},
  year = {2023},
  publisher = {Hugging Face},
  howpublished = "\url{https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard}"
}

@software{eval-harness,
  author       = {Gao, Leo and
                  Tow, Jonathan and
                  Biderman, Stella and
                  Black, Sid and
                  DiPofi, Anthony and
                  Foster, Charles and
                  Golding, Laurence and
                  Hsu, Jeffrey and
                  McDonell, Kyle and
                  Muennighoff, Niklas and
                  Phang, Jason and
                  Reynolds, Laria and
                  Tang, Eric and
                  Thite, Anish and
                  Wang, Ben and
                  Wang, Kevin and
                  Zou, Andy},
  title        = {A framework for few-shot language model evaluation},
  month        = sep,
  year         = 2021,
  publisher    = {Zenodo},
  version      = {v0.0.1},
  doi          = {10.5281/zenodo.5371628},
  url          = {https://doi.org/10.5281/zenodo.5371628}
}

@misc{clark2018think,
      title={Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge},
      author={Peter Clark and Isaac Cowhey and Oren Etzioni and Tushar Khot and Ashish Sabharwal and Carissa Schoenick and Oyvind Tafjord},
      year={2018},
      eprint={1803.05457},
      archivePrefix={arXiv},
      primaryClass={cs.AI}
}

@misc{zellers2019hellaswag,
      title={HellaSwag: Can a Machine Really Finish Your Sentence?},
      author={Rowan Zellers and Ari Holtzman and Yonatan Bisk and Ali Farhadi and Yejin Choi},
      year={2019},
      eprint={1905.07830},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

@misc{hendrycks2021measuring,
      title={Measuring Massive Multitask Language Understanding},
      author={Dan Hendrycks and Collin Burns and Steven Basart and Andy Zou and Mantas Mazeika and Dawn Song and Jacob Steinhardt},
      year={2021},
      eprint={2009.03300},
      archivePrefix={arXiv},
      primaryClass={cs.CY}
}

@misc{lin2022truthfulqa,
      title={TruthfulQA: Measuring How Models Mimic Human Falsehoods},
      author={Stephanie Lin and Jacob Hilton and Owain Evans},
      year={2022},
      eprint={2109.07958},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

@misc{DBLP:journals/corr/abs-1907-10641,
      title={{WINOGRANDE:} An Adversarial Winograd Schema Challenge at Scale},
      author={Keisuke Sakaguchi and Ronan Le Bras and Chandra Bhagavatula and Yejin Choi},
      year={2019},
      eprint={1907.10641},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

@misc{DBLP:journals/corr/abs-2110-14168,
      title={Training Verifiers to Solve Math Word Problems},
      author={Karl Cobbe and
                  Vineet Kosaraju and
                  Mohammad Bavarian and
                  Mark Chen and
                  Heewoo Jun and
                  Lukasz Kaiser and
                  Matthias Plappert and
                  Jerry Tworek and
                  Jacob Hilton and
                  Reiichiro Nakano and
                  Christopher Hesse and
                  John Schulman},
      year={2021},
      eprint={2110.14168},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	75.66
AI2 Reasoning Challenge (25-Shot)	72.78
HellaSwag (10-Shot)	88.77
MMLU (5-Shot)	64.74
TruthfulQA (0-shot)	70.89
Winogrande (5-shot)	86.42
GSM8k (5-shot)	70.36